Chinese Journal of Computers   Full Text
Title	Using Service Clustering to Facilitate Process-Oriented Semantic Web Service Discovery
Authors	SUN Ping1),2),3) JIANG Chang-Jun1),2)
Address	1)(Department of Computer Science & Engineering, Tongji University, Shanghai 201804) 2)(Key Laboratory of Embedded System and Service Computing of Ministry of Education, Tongji University, Shanghai 201804) 3)(School of Software Engineering, Tongji University, Shanghai 201804)
Year	2008
Issue	No.8(1340—1353)
Abstract &   Background	Abstract The discovery of suitable Web services for a given user requirement is one of the central operations in Service-oriented Architectures. With the popularization of Web services, researches on minimizing the discovery duration and improving precision ration are getting more important. This paper proposes a mechanism to support semantic Web service discovery. Petri net is adopted as a modeling language for the specification of service process model. On the one hand, the methodology of service clustering groups similar services according to the functional similarity and process similarity. The utilization of service clustering can potentially enable service matchmaker to significantly reduce the overhead, deploy the discovery of candidate services quickly. On the other hand, for a service request, a service matchmaker compares the functionality compatibility and process consistency with the candidate services. It is an extension of previous functionality-driving service matchmaking approaches, thus leading to more accurate matchmaking. In the end, some simulation results are demonstrated to show the effectiveness of the proposed method. Keywords service discovery; matchmaking; process model; Petri net; clustering Background The discovery of suitable Web services for a given user requirement is one of the central operations in Service-oriented Architectures. With the popularization of Web services, researches on minimizing the discovery duration and improving precision ration are getting more important. Data mining techniques, especially clustering analysis, can be used to group similar services. As a consequence, service matchmaker can reduce time and cost by matchmaking Web services in well-classified service repository. In addition, most of the work in Web service discovery has focused on the service functional description, especially IOPE (Input, Output, Precondition and Effect). To overcome the inaccuracy of service discovery, both the service profile and the process information should be taken into consideration. Unfortunately, the researches in this area have not received a lot of attention so far. This paper proposes a mechanism to support semantic Web service discovery. Petri net is adopted as a modeling language for the specification of service process model. On the one hand, the methodology of service clustering groups similar services according to the functional similarity and process similarity. The utilization of service clustering can potentially enable service matchmaker to significantly reduce the overhead, deploy the discovery of candidate services quickly. On the other hand, for a service request, a service matchmaker compares the functionality compatibility and process consistency with the candidate services. It is an extension of previous functionality-driving service matchmaking approaches, thus leading to more accurate matchmaking. This paper is supported by the National High Technology Research and Development Program(863 Program) of China under grant No.2007AA01Z136, and the National Basic Research Program(973 Program) of China under grant No.2003CB316902. Both projects aim at proposing a reliable approach to semantic Web service composition, including: service description, service discovery, service composition, and composite service verification. The research group has published several high quality papers on dynamic service composition, QoS-aware optimal service selection, in "IEEE Transactions on System, Man, and Cybernetics", "Information and Software Technology", "Journal of Software", and other respected journals both at home and abroad. This paper mainly focuses on the field of discovery of service. It utilizes service clustering and process matchmaking to potentially deploy the discovery of candidate services quickly, and lead to more accurate matchmaking.